Purchase this article with an account.
Benno Petrig, Jeff Clendenon, Lisa Ensman, Matthew Muller, Ann Elsner, Glen Ozawa, Taras Litvin, Jorge Cuadros, Tuhin Roy, Danny Li; Comparison of Fixation Stability Measured by Laser Scanning Digital Camera and DLP-Cam. Invest. Ophthalmol. Vis. Sci. 2013;54(15):2773.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To compare fixation stability in diabetic patients measured by the laser scanning digital camera (LSDC) and the digital light projector camera (DLP-Cam), using standard deviation of Euclidean distance (ED) and bivariate contour ellipse area (BCEA) as metrics.
Eight diabetic patients (5 male, 3 females, age 48±12 yr) were imaged at the Eastmont Wellness Center, a diabetic retinopathy screening site within the EyePACS telemedicine network, with both the LSDC and the DLP-Cam. LSDC and DLP-Cam have 36 and 33 deg fields, respectively. Light levels at the cornea were 2 mW @ 850 nm, which appeared quite dim, for the LSDC and 42 uW from the green LED, which was bright, for the DLP-Cam. Image buffers of 20 @ 11 fps (LSDC) and 16 @ 17 fps (DLPC) frames were acquired on the same day by the same camera operator. One eye was chosen at random for this analysis. Fixation stability was assessed using the coordinates determined for image registration by the same alignment algorithm. The measurements in pixels were converted to minArc before comparing the LSDC and DLPC results by paired t-tests.
The standard deviation of ED for this group of patients was 5.09±3.42 and 5.69±3.25 minArc for LSDC and DLP-Cam, respectively. The BCEA, which typically decreases the effect of outliers, was 2.43±0.41 and 2.43±0.44 log(minArc^2), for LSDC and DLP-Cam, respectively. The differences in fixation stability were not statistically significant by either metric.
Both imaging systems provide comparable fixation stability measures, suggesting that the brighter fundus illumination of the DLP-Cam compared to the LSDC does not affect fixation ability. Our alignment algorithms are able to process images taken with either visible or near infrared fundus illumination which are known to emphasize different retinal structures.
This PDF is available to Subscribers Only